U.S. patent application number 09/815594 was filed with the patent office on 2001-10-04 for image processing apparatus and image display apparatus using same.
Invention is credited to Furukawa, Hiroyuki, Yoshida, Yasuhiro.
Application Number | 20010026283 09/815594 |
Document ID | / |
Family ID | 26588300 |
Filed Date | 2001-10-04 |
United States Patent
Application |
20010026283 |
Kind Code |
A1 |
Yoshida, Yasuhiro ; et
al. |
October 4, 2001 |
Image processing apparatus and image display apparatus using
same
Abstract
An image processing apparatus of the present invention
comprising (a) a first signal processing circuit for applying gamma
correction to an n-bit (n: a natural number) digital signal
inputted as a video signal and for converting the n-bit digital
signal into an m-bit (m>n, m: a natural number) digital signal,
and (b) a second signal processing circuit for adding a noise
signal, which is used for pseudo contour reduction, into the m-bit
digital signal from the first signal processing circuit and for
outputting a Q-bit (Q: a natural number) digital signal, which is
obtained from rounding off a less significant (m-Q) bit
(Q.ltoreq.n) from the m-bit digital signal, to a display
section.
Inventors: |
Yoshida, Yasuhiro;
(Nara-shi, JP) ; Furukawa, Hiroyuki; (Ueno-shi,
JP) |
Correspondence
Address: |
Dike, Bronstein, Roberts & Cushman LLP
130 Water Street
Boston
MA
02109
US
|
Family ID: |
26588300 |
Appl. No.: |
09/815594 |
Filed: |
March 23, 2001 |
Current U.S.
Class: |
345/600 ;
348/E5.074; 348/E5.133 |
Current CPC
Class: |
G09G 2320/0276 20130101;
G09G 3/20 20130101; H04N 5/66 20130101; G09G 3/2011 20130101; G09G
2320/0626 20130101; G09G 2320/0285 20130101; G09G 2320/0673
20130101; G09G 3/3648 20130101; G09G 3/2092 20130101; H04N 5/202
20130101; G09G 2320/0242 20130101; G09G 2320/0606 20130101; G09G
3/2059 20130101; G09G 3/3614 20130101; G09G 3/2048 20130101; G09G
2320/02 20130101 |
Class at
Publication: |
345/600 |
International
Class: |
G09G 005/02 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 24, 2000 |
JP |
2000-84663 |
Jan 22, 2001 |
JP |
2001-13864 |
Claims
What is claimed is:
1. An image processing apparatus, comprising: a first signal
processing circuit for applying gamma correction to an n-bit (n: a
natural number) digital signal inputted as a video signal, and for
converting the digital signal into an m-bit (m>n, m: a natural
number) digital signal; and a second signal processing circuit for
adding a noise signal to the m-bit digital signal from said first
signal processing circuit, and for outputting a Q-bit (Q: a natural
number) digital signal obtained from rounding off a less
significant (m - Q) bit (Q <n) from the m-bit digital
signal.
2. The image processing apparatus as set forth in claim 1, wherein:
said first signal processing circuit includes bit converting means
for converting the inputted n-bit digital signal into the m-bit
digital signal in accordance with a pre-set value.
3. The image processing apparatus as set forth in claim 2, wherein:
said bit converting means is a Look Up Table which outputs the
m-bit digital signal that is the present value in accordance with
the inputted n-bit digital signal.
4. The image processing apparatus as set forth in claim 2, wherein:
said bit converting means is a calculating device for converting
the n-bit digital signal into the m-bit (m>n) digital signal by
numerical calculation.
5. The image processing apparatus as set forth in claim 1, wherein:
said first signal processing circuit and said second signal
processing circuit are provided for respective RGB colors.
6. The image processing apparatus as set forth in claim 1, wherein:
an average value of a signal level of the noise signal is set to
zero.
7. The image processing apparatus as set forth in claim 1, wherein:
the noise signal is a random noise signal with no regularity in its
cycle of amplitude.
8. The image processing apparatus as set forth in claim 1, wherein:
the noise signal is obtained from, by using an arbitrary noise
pattern table, switching a starting point of the noise pattern
table per field or per noise pattern table.
9. The image processing apparatus as set forth in claim 1, wherein:
a histogram of an amplitude of the noise signal shows Gaussian
dispersion where zero amplitude of the noise signal is at the
center.
10. An image processing apparatus, comprising: a signal processing
circuit for adding a noise signal to an inputted m-bit (m: a
natural number) digital signal, and for outputting a Q-bit (Q: a
natural number) digital signal obtained from rounding off a less
significant (m-Q) bit (Q<m) from the m-bit digital signal.
11. An image display apparatus comprising a display means for
displaying an image, and driving means for driving the display
means, said image display apparatus, further comprising: an image
processing apparatus that includes (a) a first signal processing
circuit for applying gamma correction to an n-bit (n: a natural
number) digital signal inputted as a video signal and for
converting the n-bit digital signal into an m-bit (m>n, m: a
natural number) digital signal, and (b) a second signal processing
circuit for adding a noise signal to the m-bit digital signal from
said first signal processing circuit and for outputting a Q-bit (Q:
a natural number) digital signal obtained from rounding off a less
significant (m-Q) bit (Q.ltoreq.n) from the m-bit digital
signal.
12. The image display apparatus as set forth in claim 11, wherein:
said first signal processing circuit includes bit converting means
for converting the inputted n-bit digital signal into the m-bit
digital signal in accordance with a pre-set value.
13. The image display apparatus as set forth in claim 12, wherein:
the pre-set value in said bit converting means is rewritable so
that unevenness in properties of said driving means may be
absorbed.
14. The image display apparatus as set forth in claim 12, wherein:
the pre-set value in said bit converting means is rewritten in
accordance with brightness in surroundings of said image display
apparatus.
15. The image display apparatus as set forth in claim 12, wherein:
the pre-set value in said bit converting means is rewritten in
accordance with brightness of overall display image of said display
means.
16. An image display apparatus, comprising: an image processing
apparatus which includes a signal processing circuit for adding a
noise signal to an inputted m-bit (m: a natural number) digital
signal and for outputting a Q-bit (Q: a natural number) digital
signal obtained from rounding off a less significant (m-Q) bit
(Q<m) from the m-bit digital signal.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to an image processing
apparatus, in which a digital signal is inputted as a video signal,
for use in a digital image display apparatus for displaying a
digital image, and an image display apparatus provided with the
image processing apparatus.
BACKGROUND OF THE INVENTION
[0002] Conventionally, gamma correction is usually applied to a
digital video signal inputted into a digital image display
apparatus in order to improve display properties to the level of
that of Cathode Ray Tube (CRT) apparatus.
[0003] However, pseudo contour is caused on the display image when
inputting the gamma-corrected video signal into the digital image
display apparatus without any treatment, thus resulting in lower
display quality.
[0004] Therefore, for example, disclosed in Japanese Unexamined
Patent Publication Tokukaihei No.9-185707 (published on Jul. 15,
1997) is technology for reducing the pseudo contour due to the
gamma correction. The technology employs signal processing
technology in which error diffusion method is applied, in other
words, the gamma correction, which is a vital process for a spatial
light modulator and a display apparatus with linear luminescence
properties, is carried out by random addition of a controlled noise
signal into the video signal.
[0005] However, in the error diffusion method disclosed in the
publication, the noise signal, which is to be added to the video
signal, is stored and fed back in order to prevent accumulation of
errors. The apparatus is complicated due to the additional need for
a memory, which is used for storing the noise signal, and for a
circuit to feed back. Consequently, a high cost is a problem.
SUMMARY OF THE INVENTION
[0006] An object of the present invention is to provide, at a low
cost, an image processing apparatus that may achieve reduction of
pseudo contour, which is generated by gamma correction, by using a
simple circuit structure, and to provide an image display apparatus
provided with the image processing apparatus.
[0007] The image processing apparatus of the present invention is,
for the purpose of achieving the object, utilized for an image
display apparatus that is provided with display apparatus having
display properties of n bit (n: a natural number) for displaying an
image, when accepting a digital signal as a video signal, in
accordance with the digital signal. The image processing apparatus
is provided with a first signal processing circuit and a second
signal processing circuit. The first signal processing circuit
carries out the gamma correction for an n-bit digital signal
inputted as the video signal, and converts the digital signal into
an m-bit (m>n, m: a natural number) digital signal. The second
signal processing circuit adds a noise signal, which reduces the
pseudo contour, to the m-bit digital signal from the first signal
processing circuit, then outputs a Q-bit (Q: a natural number)
digital signal, which is obtained by rounding off a less
significant (m-Q) bit (Q.ltoreq.n) from the m-bit digital signal,
to the above display apparatus.
[0008] Therefore, the inputted n-bit digital signal is expanded to
the m-bit digital signal after the gamma correction. Next, the
noise signal for reducing the pseudo contour is added to the m-bit
digital signal, which is converted into the Q-bit digital signal by
rounding off the less significant (m - Q) bit from the m-bit
digital signal. Then, the n-bit digital signal is inputted into the
display apparatus of Q-bit display properties.
[0009] Hence, the digital signal without the bit lack is inputted
into the display apparatus. In other words, the display apparatus
having 8-bit display properties receives the 8-bit digital
signal.
[0010] Moreover, a radical change in thickness of the color of
adjacent pixels to be displayed is prevented by adding the noise
signal, which reduces the pseudo contour, into the m-bit digital
signal. The m-bit digital signal has become a video signal with no
or little pseudo contour generation.
[0011] Next, the less significant (m-Q) bit of the m-bit digital
signal with no or little pseudo contour generation is rounded off
to convert the m-bit digital signal into the Q-bit digital signal
whose bit number is the same as the display properties of the
display apparatus. The Q-bit digital signal has become a video
signal in which the thickness of the color of the adjacent pixels
of the image to be displayed is not radically changed, in other
words, no pseudo contour is generated.
[0012] Therefore, a display image of high quality with no or little
pseudo contour may be achieved by inputting this Q-bit digital
signal into the display apparatus having Q-bit display
properties.
[0013] In order to achieve the above object, the image processing
apparatus of the present invention is provided with a signal
processing circuit for outputting the Q-bit digital signal, which
is obtained by rounding off the less significant (m-Q) bit (Q<m,
m & Q: natural numbers) from the m-bit digital signal after
adding the noise signal for reducing the pseudo contour into the
inputted m-bit digital signal.
[0014] Because the less significant (m-Q) bit is rounded off after
the addition of the noise signal into the m-bit digital signal, it
is not the thin out of the less significant (m-Q) bit. Thus, it is
possible that the display properties equivalent to m bit are
expressed in a pseudo manner by the Q-bit digital signal.
[0015] Set to zero is the average value of the signal level of the
noise signal to be added into the m-bit digital signal.
Accumulation of errors of signals due to the noise signal addition
will not be caused by adding a noise signal, which is in the above
setting, into the digital signal in a random manner.
[0016] This may simplify the apparatus, thus offers a low-cost
image processing apparatus, because some apparatus such as a memory
and a feedback circuit are no longer necessary, which are needed in
the conventional error diffusion method for preventing the
accumulation of errors when a noise signal is added into a digital
signal.
[0017] For a fuller understanding of the nature and advantages of
the invention, reference should be made to the ensuing detailed
description taken in conjunction with the accompanying
drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] FIG. 1 is a schematic view showing a schematic structure of
an image display apparatus provided with an image processing
apparatus of the present invention.
[0019] FIG. 2 (a) is a schematic view showing a schematic structure
of a first signal processing section provided in the image
processing apparatus shown in FIG. 1.
[0020] FIG. 2 (b) is a graph illustrating a relation between an
input value and an output value in the first signal processing
section shown in FIG. 2 (a).
[0021] FIG. 3 is a graph showing a relation between a signal level
and transmittance of the input signal of the first signal
processing section shown in FIG. 2.
[0022] FIG. 4 is a graph showing display properties of a Liquid
Crystal Display (LCD) of a display section provided to the image
display apparatus shown in FIG. 1.
[0023] FIG. 5 is a graph illustrating a relation between a signal
level and transmittance of the output signal of the first signal
processing section shown in FIG. 2.
[0024] FIG. 6 is a schematic view showing a schematic structure of
a second signal processing section provided in the image processing
apparatus shown in FIG. 1.
[0025] FIG. 7 is a schematic view showing a schematic structure of
a noise generating circuit provided to the second signal processing
section shown in FIG. 6.
[0026] FIG. 8 is an explanatory view showing a random noise
signal.
[0027] FIG. 9 is a histogram showing frequency distribution of
levels of the noise signal shown in FIG. 8.
[0028] FIGS. 10 (a) to (d) are graphs showing relations between the
input signal and the output signals of the signal processing
section shown in FIG. 1 in case of color image display. FIG. 10 (a)
is a graph illustrating a relation between a signal level of the
input signal and the transmittance of the LCD. FIG. 10 (b) is a
graph explaining a relation between a signal level of an R (Red)
signal among the output signals and the transmittance of the LCD.
FIG. 10 (c) is a graph showing a relation between a signal level of
a G (Green) signal among the output signals and the transmittance
of the LCD. FIG. 10 (d) is a graph illustrating a relation between
a signal level of a B (Blue) signal among the output signals and
the transmittance of the LCD.
[0029] FIG. 11 is a schematic diagram showing an image display
apparatus for showing connection formation of respective circuits
for rewriting a value of the LUT of the image processing apparatus
of the present invention.
[0030] FIG. 12 is an explanatory diagram showing how a video signal
and a synchronizing signal fed into driving IC, which is used by a
display section of the image display apparatus shown in FIG. 11,
are inputted.
DESCRIPTION OF THE EMBODIMENTS
[0031] Described below is the explanation of one embodiment of the
present invention. It should be noted that a Liquid Crystal Display
(LCD) apparatus will be discussed as the image display apparatus in
the present embodiment.
[0032] The LCD apparatus in the present embodiment is, as shown in
FIG. 1, provided with a display section 11 as display means for
displaying an image in accordance with a video signal, and an image
processing apparatus 12 for processing the video signal in
accordance with the display properties of the display section
11.
[0033] The display section 11 includes an LCD 13, which has display
properties of Q bit (Q: a natural number), that is, has tone
gradation of 2.sup.Q (the Qth power of 2), and a source driver 14
and a gate driver 15 as drive means for driving the LCD 13.
[0034] The source driver 14, upon receipt of a video signal that
has been processed by the image processing apparatus 12, sends a
voltage, which varies depending on the inputted video signal, to a
source electrode line (not shown) of the LCD 13.
[0035] On the other hand, the gate driver 15, upon receipt of
synchronizing signals (Horizontal Synchronizing Signal H, Vertical
Synchronizing Signal V) that have been outputted from a
synchronizing signal generating circuit (not shown), sends a
voltage, which varies depending on the inputted synchronizing
signals, to a gate electrode line (not shown).
[0036] The image processing apparatus 12 is provided with a first
signal processing section (a First Signal Processing Circuit) 16
and a second signal processing section (a Second Signal Processing
Circuit) 17. The first signal processing section 16 has bit
converting means that applies gamma correction to an n-bit (n: a
natural number) digital signal inputted as a video signal, and
converts the digital signal into an m-bit (m>n, m: a natural
number) digital signal, then outputs the digital signal. The second
signal processing section 17 converts the m-bit digital signal,
which is forwarded from the first signal processing section 16,
into a Q-bit (Q: a natural number) digital signal by rounding down
the m bit below a less significant (m-Q) bit (Q.ltoreq.n).
[0037] Therefore, the bit number Q of the digital signal inputted
into the source driver 14 of the display section 11 is equal or
less than n-bit number that is inputted in.
[0038] As the result, the display quality of display image may be
significantly improved in the LCD apparatus in the above-mentioned
arrangement by avoiding pseudo contour generation, which is caused
by the bit lack occurred at the gamma correction of a video
signal.
[0039] In the following explanation, the values of those natural
numbers, namely n, m, and Q, are supposed to be n=8, m=10 and Q=8,
respectively. Thus, the LCD 13 will be described as a display
apparatus with display properties of Q=8 bits. In the image
processing apparatus 12, it is supposed that the first signal
processing section 16 converts an n-bit, that is, 8-bit digital
signal into an m-bit, that is, 10-bit digital signal by expanding
bit numbers from 8 (N) to 10 (M), then the second signal processing
section 17 converts the m-bit, that is, 10-bit digital signal to a
Q-bit, that is, 8-bit digital signal by rounding off less
significant 2 bits (m-Q) from the m-bit digital signal.
[0040] The image processing apparatus 12 is discussed below.
[0041] To begin with, the signal processing made by the first
signal processing section 16 in the image processing apparatus 12
is explained below, referring to FIG. 2 through FIG. 5. Note that,
the LCD 13 of the display section 11 is made of liquid crystals of
normally white mode, in which its transmittance is decreased as the
applied voltage is increased at the display electrode.
[0042] The first signal processing section 16 is, as shown in FIG.
2 (a), provided with a Look Up Table (LUT) 18 as bit converting
means to carry out the gamma correction of an inputted 8-bit
digital signal and the conversion of the digital signal into a
10-bit digital signal. In other words, the LUT 18 is, as shown in
FIG. 2 (b), provided for converting the signal level (input value)
of the 8-bit digital signal into the corresponding signal level
(output value) of the 10-bit digital signal.
[0043] The setting of the output value of the LUT 18 is explained
below, with reference to FIG. 3 through FIG. 5.
[0044] FIG. 3 is a graph illustrating a relation between the signal
level of the inputted 8-bit digital signal and the transmittance of
the LCD 13 under the condition of gamma=2.2, which is commonly used
for television systems. FIG. 4 is a graph showing a relation
between the voltage applied to the electrode of the LCD 13 and the
transmittance of the LCD 13 in order to explain the display
properties of the LCD 13. Moreover, FIG. 5 is a graph showing a
relation between the transmittance of the LCD 13 and the signal
level of the digital signal when the 8-bit digital signal is
converted into the 10-bit digital signal as shown in FIG. 2
(b).
[0045] The graph in FIG. 5 is prepared in the following manner.
According to FIG. 4, the transmittance is 100% at applied voltage
of 1.00 V. The signal level of the inputted digital signal in this
condition is 1023, while the signal level of the inputted digital
signal will be lowered to 0 when the transmittance is 0% at applied
voltage of 4.50 V. Similarly, the transmittance of 50% at 2.50 V
applied voltage gives a signal level of 584, which is calculated
from a formula: (1023-(2.50-1.00)/(4.50-1.00)- .times.1024).
Accordingly, the signal level will be 379 with 20% transmittance at
3.20 V applied voltage, 672 with 70% transmittance at 2.20 V
applied voltage, and 818 with 90% transmittance at 1.70 V applied
voltage.
[0046] Here, the signal level is indicative of the tone gradation
of the video signal. Therefore, the scale of axis of abscissa in
FIG. 3 is 0 to 255 because 8 bits have a gradation of 2.sup.8=256,
while that scale in FIG. 5 is 0 to 1023 since the gradation is
2.sup.10=1024 for 10 bits.
[0047] FIG. 2 (b) is provided for correlating FIG. 5 and FIG. 3
obtained in the manner mentioned above. For example, with 50%
transmittance, the input level is 192 according to FIG. 3, while
the value for 10 bits is 584 according to FIG. 5. Thus, FIG. 2 (b)
is the graph for converting the input with the input value of 192
to the output value of 584. Similarly, the graph shows the
conversion of other values as follows: from input value 128 to
output value 379 at 20% transmittance, from input value 220 to
output value 672 at 70% transmittance and from input value 242 to
output value 818 at 90% transmittance.
[0048] FIG. 2 (b) is prepared by calculating the correlation for
all the levels in this manner.
[0049] The digital signal, which was inputted in 8 bits, is
converted into the 10-bit digital signal by using the LUT 18 where
its output value has been determined in the way discussed above. A
relation between the signal level of the 10-bit digital signal
after the conversion and the transmittance of the LCD 13 gives the
graph shown in FIG. 5.
[0050] Described below is an explanation on the signal processing
of the second signal processing section 17 in the image processing
apparatus 12.
[0051] As shown in FIG. 6, the second signal processing section 17
is provided with a Bit-Depth Extension (BDE) 22 that is composed of
a noise generating circuit 19 for generating a noise signal to be
added into an inputted digital signal, an adding circuit 20 for
adding the noise into the inputted digital signal forwarded from
the noise generating circuit 19, and a bit number converting
circuit 21 for converting the bit number of the digital signal in
which the noise signal is added. Synchronizing signals (Horizontal
Synchronizing Signal H, Vertical Synchronizing Signal V), which are
outputted from a synchronizing signal generating circuit (not
shown), are inputted into the noise generating circuit 19.
[0052] Thus, the second signal processing section 17 converts the
10-bit digital signal, which is outputted from the first signal
processing section 16, into an 8-bit digital signal by adding the
noise signal and rounding off its less significant 2 bits from the
10-bit digital signal.
[0053] In the second signal processing section 17, by adding the
noise signal into the digital signal for reducing the pseudo
contour, it can be avoided to radically change the thickness of
color of pixels adjacent to the pixels on which the image is to be
displayed. Thus, the 10-bit digital signal with the noise signal
addition has become a video signal with rare occurrence of the
pseudo contour.
[0054] Next, the 10-bit digital signal with the rare occurrence of
the pseudo contour is converted into an 8-bit digital signal after
rounding off its less significant 2 bits from the 10-bit digital
signal, then inputted into the display section 11 with 8-bit
display properties. The 8-bit digital signal, which is inputted
into the display section 11, has become a video signal without a
radical change in the thickness of color of the pixels adjacent to
the pixels on which the image is to be displayed. In short, it has
become a video signal free from the pseudo contour generation.
[0055] Therefore, a display image of high quality with no or little
pseudo contour can be achieved by inputting the 8-bit digital
signal into the display means with 8-bit display properties.
[0056] The noise generating circuit 19 is provided for producing a
noise signal to prevent the generation of the pseudo contour in the
video signal.
[0057] The average value of the signal level of the generated noise
signal is set to zero.
[0058] Alternatively, a noise can be generated by using a specific
pattern table, as described below. The noise generating circuit 19
is composed of a noise ROM 41 and an address counter 42, as
illustrated in FIG. 7. The address counter 42 has received the
horizontal synchronizing signal H, the vertical synchronizing
signal V, and clock (not shown).
[0059] A noise enough for one screen is stored in the noise ROM 41,
while a horizontal address is incremented for each horizontal
pixel, and a vertical address is incremented for each vertical line
in the address counter 42, where its output is added to the noise
ROM 41. As the result, a noise for one screen is consecutively
supplied per pixel from the noise ROM 41. The noise is added to the
video signal by the adding circuit 20 shown in FIG. 6.
[0060] The noise ROM 41 may be arranged so that the noise is stored
per small block such as 16.times.16 or 32.times.32 size, instead of
one screen. In this case, the address counter 42 is adapted for
being reset every 16 or 32 in order that the same noise data are
repeatedly outputted per block. The horizontal synchronizing signal
H and the vertical synchronizing signal V, which are added to the
address counter 42, are used for this purpose. In addition, a
problem that periodicity is visualized will not occur even when the
noise data are dealt per block such as 16 or 32, because the noise
data are a random signal as explained later. The above arrangement
gives such an advantage that ROM capacity of the noise ROM 41 may
be significantly cut down.
[0061] Moreover, the address counter 42 may be adapted as described
below. In short, the section may be adapted so that leading address
is altered for each field of a screen (each vertical synchronizing
signal V) or for each block border.
[0062] For example, when a noise is transformed into a 16.times.16
block structure, where the read-out of noise is carried out
consecutively per pixel, in a first block, the read-out starts from
address (0,0) of the noise ROM 41 and is continued until address
(15,15). In a following block, it is designed that the read-out is
performed from address (1,1) of the noise ROM 41 to address
(15,15), then, through addresses (0,0) and (0,15), continued until
address (1,0). In a next block, starting from address (2,2), the
read-out is continued until address (15,15), then, via addresses
(0,0), (0,15), (1,0) and (1,15), ends at address (2,1). The
remaining will be read out in the same fashion.
[0063] In this example, the starting addresses of both of the
horizontal and the vertical ones, are incremented by one address
per block border, in other words, for each read-out of one block.
But it is not limited to this example. For example, more than two
addresses may be incremented, or either of the horizontal or the
vertical one may be incremented.
[0064] Alternatively, when the leading addresses are changed per
frame, where the read-out of noise is carried out consecutively per
pixel, in the first frame, the read-out starts from address (0,0)
of the noise ROM 41 and is continued until address (15,15).
Similarly, in the following frame, the read-out is also carried out
consecutively from address (0,0). The readout in this frame is
repeated consecutively in the subsequent series of read-outs. In
the next frame, it is designed that the read-out, starting from
address (1,1) of the noise ROM 41, is continued until address
(15,15), then, through addresses (0,0) and (0,15), continued until
address (1,0). The read-out in this frame is continued in this
manner until the end. In the further next frame, starting from
address (2,2), the read-out is continued until address (15,15),
then via addresses (0,0), (0,15), (1,0) and (1,15), is ended at
address (2,1). The remaining will be read out in the same
manner.
[0065] In this case, as to both of the horizontal and the vertical
addresses, the starting addresses are incremented by one address
per frame. But it is not limited to this case. For example more
than two addresses may be incremented, or either of the horizontal
or the vertical may be incremented.
[0066] Discussed above are the methods for performing the read-out
per block in the frame and altering the starting address, and the
method for carrying out the read-out per frame and altering the
starting address. However, other similar combinations may be
thought of for convenience. The block border, which is originally
inconspicuous, may be made further inconspicuous by altering the
read-out starting address consecutively in the above manner,
because the block structure of the noise, which has a block
structure such as 16.times.16, is vanished by using this kind of
read-out method.
[0067] Moreover, especially for the method in which the starting
position is altered per frame, the following specific effects are
produced. An LCD apparatus is generally unable to respond to a
speed faster than a speed corresponding to one frame, thus cannot
perfectly follow a noise changed per frame. As a result, the value
obtained from the sum of a signal value and the average value of
the noise is displayed per pixel. For this reason, errors, which
are equivalent to those errors when all pixels are rounded down
evenly, may be generated in this method, even with a process for
rounding down a signal by some bits as discussed later. This means
that all pixels may express all the quantization levels.
[0068] On the contrary, in the method previously explained where a
noise is not altered per frame, the tone gradation is expressed in
so-called the principle of area tone gradation. Therefore, when the
respective pixels are considered individually, it cannot be said
that all the quantization levels are expressed for the respective
pixels. Thus, the method may be called as a method to express all
the tone gradations on average in accordance with the size of a
screen.
[0069] Accordingly, in the method in which a noise is changed per
frame, all the tone gradations may be expressed per pixel. If this
method is applied together with a method where leading addresses
are altered per block, as described previously, it produces some
properties in which the block border is inconspicuous even with
little memory capacity. Furthermore, as discussed later, extra
properties that are suitable for liquid crystals may also be
created by devising the properties of the noise.
[0070] This prevents the accumulation of average errors in the
digital signal, to which the noise signal generated by the noise
generating circuit 19 is added, and eliminates the need of a
special circuit for the prevention of the error accumulation.
[0071] Note that, to make zero the average value of the signal
level in the generated noise signal may be easily achieved, for
example, by utilizing random number generating software and
algorithm by which the average of the amplitude of the signal of
the generated random numbers may become zero.
[0072] The random number generating software is used for generating
a random noise signal. The noise signal preferably has a low
amplitude auto-correlation between adjacent pixels, and has no
amplitude value of zero in its dispersion range. Also, the
histogram of the amplitude of the noise signal preferably shows
Gaussian dispersion in which the zero amplitude of the noise signal
is at the center.
[0073] Therefore, the random noise signal has such characteristics
that, as shown in FIG. 8, the amplitude of noise level .DELTA.V
will never exceed the upper and lower limits of the level and the
average value of the level is always zero within a predetermined
range (for example, one field). The histogram of noise level is, as
shown in FIG. 9, regarded that it is in its most natural state when
it shows the Gaussian dispersion.
[0074] If the noise signal has low auto-correlation between
adjacent pixels, the noise signal to be added has no periodicity
but high randomness in the cycle of its amplitude. This is the
point where the present invention is effective especially when used
for an LCD apparatus. Dot reversing, line reversing or frame
reversing drives are generally carried out in an LCD apparatus in
order to perform AC drive. However, different LCDs have different
reversing cycles. Thus, the cycle of reversing is not
uniformed.
[0075] Those reversing drives cause killer pattern in LCDs. In a
rare case, it is required to display the killer pattern, and a
significant degradation in image quality will be caused. Therefore,
not all images can be displayed in a satisfactory image quality. In
addition, because the reversing drive is carried out based on an
artificial regular cycle, the killer-pattern-affected image is
likely given a regular cycle.
[0076] In a rare case, its regularity is interfered with that of
the reversing drive, and an added noise signal with regular cycles
of the amplitude may cause a significant image degradation. For
instance, while an LCD apparatus has no such problem, other LCD
apparatus may be suffered from the problem. For most of such cases,
the problem is so severe that not only image quality is difficult
to be improved but also the cause is hard to be found.
[0077] On the contrary, use of an irregular noise signal as the
added signal like the present invention causes no interference with
the regularity of the reversing drive, thus causing no image
degradation. In addition, no killer pattern is produced for any LCD
apparatus, thus causing no image degradation.
[0078] Moreover, several advantages are given by the noise signal
with zero value of the average value of the amplitude, and
explained below. In this circuit, since noise signals with
different values should be added to each pixel in order to reduce
the pseudo contour, it is required to change the values
corresponding to the added noise signals for each pixel. But, it
will be a problem that the brightness of the screen is changed for
a large area. The brightness of the screen for such a large area
changes by the amount corresponding to the average value of the
amplitude of the noise signal when adding of the noise signal.
Accordingly, the average value of the noise signal should be zero
in order that the brightness may not change during the addition of
the noise signal. Therefore, only the reduction of the pseudo
contour is achieved by setting to zero the average value of the
amplitude of the noise signal, without undesirable changes in the
brightness in a large area of the screen.
[0079] Furthermore, since the noise signal has a histogram in which
no zero value of the amplitude exists in the dispersion range and
which is the Gaussian dispersion with the zero value at the center,
the pseudo contour may be vanished irrespective of conditions even
when a video signal is rounded down to certain bits. On the other
hand, there are some cases that the zero value of the amplitude may
cancel out the effect of noise signal when the value is rounded
down.
[0080] Another advantage is that the noise of the display image
becomes least visible when the histogram of the amplitude of the
noise signal is in Gaussian dispersion. On the contrary, without
the Gaussian dispersion, the visibility of the noise itself
degrades the image quality.
[0081] Furthermore, an image processing apparatus with low cost may
be provided due to the simplified structure by cutting down a
memory and a feedback circuit, needed in the conventional error
dispersion method, for preventing the error accumulation during the
addition of a noise signal into a digital signal.
[0082] The 8-bit digital signal outputted from the second signal
processing section 17 is inputted into the source driver 14 of the
display section 11 shown in FIG. 1. Then, after conversion of
voltage into a predetermined level by the source driver 14, the
digital signal is sent to a signal line in the LCD 13.
[0083] Thus, the pseudo contour due to the bit lack will be
eliminated since the 8-bit video signal is inputted into the
display section 11 having 8-bit display properties without the bit
lack. Furthermore, the pseudo contour generated by gamma correction
is also reduced by adding a noise signal during the process in
order to reduce the pseudo contour. As the result, an image with
high display quality is achieved.
[0084] Note that, in the LCD apparatus in the above arrangement, it
is also possible to have three sets of the first signal processing
section 16 and the second signal processing section 17 shown in
FIG. 1 for RGB colors when color images are displayed.
[0085] For instance, when a video signal with such signal level
shown in FIG. 10 (a) is inputted into the first signal processing
sections 16, the LUTs 18 provided for the respective RGB colors in
the first signal processing sections 16 give signals in accordance
with the respective RGB colors as shown in FIG. 10 (b) to FIG. 10
(d). The LUTs 18 may be adapted to carry out the gamma correction
so that, when the input signal is an 8-bit signal, the output
signal is a 10-bit signal, just like the case described previously.
It is expected that there are some differences among the properties
of the respective RGB colors.
[0086] In this case, the gamma-corrected output signals of the
respective RGB colors are inputted into the second signal
processing sections 17, and outputted as 8-bit signals into the
display sections 11 after the bit conversion process.
[0087] Moreover, because the display quality of the LCD apparatus
is adjustable by the image processing apparatus 12, the image
processing apparatus 12 and the display sections 11 may be
separately provided. In this case, even with some differences among
the display sections 11, because the display quality can be
adjusted properly by the image processing apparatus 12, any types
of display sections 11 may be used for achieving an image of high
display quality.
[0088] For example, when the LCD apparatus having the above
arrangement is adapted to a display apparatus such as a personal
computer (PC) which controls the display by a video card, this may
improve the image quality by using an existing PC with a low-cost
video card only, but not by using an expensive video card, because
the high quality image is achieved not by the video card but by the
image processing apparatus 12 provided to the LCD apparatus.
[0089] By the way, the LUT 18 included in the image processing
apparatus 12 having the above arrangement, as described above, is
arranged in accordance with the display properties of the display
section 11. But, it is also possible that the LUT 18 is adapted to
use a present value or to be rewritable in real time as described
below.
[0090] Described below is the explanation on the method of
rewriting the LUT 18, giving concrete examples of the LUT 18
rewriting methods.
[0091] 1. rewriting in accordance with the display properties (V-T
curve) of the LCD.
[0092] 2. rewriting so as to absorb the unevenness between the
respective ICs (source).
[0093] 3. rewriting the bit distribution in accordance with the
brightness in surroundings.
[0094] 4. rewriting the bit distribution in accordance with the
average level of an input signal.
[0095] 5. preparing three LUTs 18 for the respective RGB colors,
and rewriting them with adjustment for the respective colors.
[0096] The apparatus for realizing the above methods is something
like an image display apparatus shown in FIG. 11. Note that,
identical members with the ones used in the previous explanation
are assigned, thus their explanation is omitted here.
[0097] The image display apparatus mentioned above is provided
with, as shown in FIG. 11, a display section 11 and an image
processing apparatus 32 that includes an LUT 18, a BDE 22, a
control circuit 23, an area judging circuit 24, an average value
calculating circuit 25 and a memory section 26. The display section
11 has the same arrangement as the image display apparatus shown in
FIG. 1.
[0098] The area judging circuit 24 is adapted to receive a
synchronizing signal identical with the one inputted into the
source driver 14 and the gate driver 15, and to judge the area to
be displayed on the LCD 13. Then, the area judging circuit 24 is
adapted to output the judged area information to the control
circuit 23.
[0099] The average value calculating circuit 25 is adapted to
receive a video signal identical with the one inputted into the LUT
18, and to calculate the average value of the brightness included
in the video signal. Then, the average value calculating circuit 25
is adapted to output the calculated value to the control circuit
23.
[0100] Moreover, the image display apparatus is provided with a
brightness sensor 27 for detecting the brightness in the
surroundings of the image display apparatus and a user's adjusting
circuit 28 for allowing a user to alter the LUT 18. Signals from
the respective sensor and circuit are inputted into the control
circuit 23.
[0101] Furthermore, a signal can be inputted from a PC 41 through
an interface 40. In this case, it is an advantage that a user is
allowed to give instructions from the PC 41.
[0102] As described, the control circuit 23 in the image processing
apparatus 32 is adapted to receive signals from the area judging
circuit 24, the average value calculating circuit 25, the
brightness sensor 27, the user's adjusting circuit 28, and the
interface 40, and to calculate the LUT 18 separately based on the
respective signals. The calculated value is outputted to the memory
section 26 connected with the control circuit 23. Note that, each
input may include all of or only part of them.
[0103] The memory section 26 is provided with a ROM and a RAM.
Written in advance in the ROM is a value for correcting the V-T
curve of the LCD 13 in accordance with the display properties of
the display section 11, because current data cannot be written into
the ROM. The RAM is adapted for storing the value from the control
circuit 23 since it is data rewritable.
[0104] If necessary, the stored value for rewriting the LUT 18 is
outputted as the LUT 18 from the memory section 26 by the control
circuit 23. This makes the LUT 18 rewritable in various
conditions.
[0105] The high quality of display image is always maintained by
rewriting the LUT 18 in accordance with the various conditions as
explained above.
[0106] Described below is the explanation about the rewriting
method of the LUT 18 for cancelling out the unevenness among the
respective ICs, with reference to FIG. 11 and FIG. 12.
[0107] As shown in FIG. 12, the source driver 14 is composed of a
plurality of source driving ICs 29, while the gate driver 15
includes a plurality of gate driving ICs 30.
[0108] The source driving ICs 29 are connected with a predetermined
number of source signal lines (not shown) of the LCD 13, while the
gate driving ICs 30 are connected with a predetermined number of
gate signal lines (not shown) of the LCD 13. For instance, when the
LCD 13 has 800.times.600 resolution and is provided with eight of
the source driving ICs 29 and six of the gate driving ICs 30, the
number of source signal lines connected to one source driving IC 29
is one hundred, while the number of gate signal lines connected to
one gate driving IC 30 is also one hundred.
[0109] Further, a synchronizing signal is inputted into the source
driving ICs 29 and the gate driving ICs 30. Hence, detected by the
synchronizing signal is the area surrounded with the respective
source driving ICs 29 and gate driving ICs 30, in other words, an
area 13a surrounded with 100.times.100 signal lines.
[0110] Those areas are driven by specific driving ICs, and have a
problem that unevenness in the properties of driving ICs cause, for
example, perceptible differences in luminance among the respective
areas. The differences in luminance among the respective blocks, in
a sense, are equivalent to the pseudo contour discussed in the
present invention. Therefore, the unevenness in screen display
properties, which are originally due to the unevenness in the
driving ICs, may be reduced by using the present invention that is
effective to suppress the pseudo contour. Moreover, the unevenness
in the properties of the respective areas can be completely
suppressed in such an arrangement that the unevenness in the
properties of the respective areas 13a of the LCD 13 are detected
in advance.
[0111] In the areas 13a detected by using a synchronizing signal as
described above, the display properties are checked in advance by a
shipping-check apparatus, in order to detect the unevenness in the
properties of the driving ICs (the source driving ICs 29 and the
gate driving ICs 30). The unevenness in the driving ICs is stored
in the memory section 26. Then, the LUT 18 is rewritten so that the
unevenness in the drive ICs is absorbed. This absorbs the
unevenness in the display properties among the respective display
areas generated in the display section 11, thus absorbing the
unevenness in the display properties of the respective display
sections 11. Thus, the unevenness between each image display
apparatus can be eliminated.
[0112] In practice, written into the LUT 18 is a value in
accordance with the properties of the source driving ICs 29
corresponding to the area of the display section 11, the area being
judged by the area judging circuit 24. In this case, the LUT 18 is
consecutively rewritten in time series since the video signal is
inputted in time series into the source operative ICs 29. The
display quality of the display image of the display section 11 is
improved by consecutively rewriting the LUT 18 in this manner.
[0113] Note that, it is also possibly arranged that the LUTs 18 are
provided in the number as many as the source driving ICs 29, and
values in accordance with the respective source driving ICs 29 are
written into the LUTs 18 in advance. In this case, it is also
possible to provide the LUTs 18 themselves in the source driving
ICs 29.
[0114] Moreover, the display image differently comes in sight
depending on the surroundings during the image displaying. In this
case, the LUT 18 may be rewritten so that, when, for example the
image display apparatus is placed in a room, the surroundings of
the image display apparatus are bright, the tone gradation of the
bright part of the image displayed on the display section 11 is
made to be high, and when the surroundings are dark, the tone
gradation of the dark part of the image displayed on the display
section 11 is made to be high. In other words, the LUT 18 may be
rewritten so that the bit distribution of the display image is
rewritten in accordance with the brightness in the surroundings of
the image display apparatus.
[0115] In practice, it may be arranged that the brightness in the
surroundings in which the image display apparatus is placed is
detected by utilizing the brightness sensor 27. Then, in accordance
with the detection signal, the control circuit 13 sets a suitable
value and then rewrites the LUT 18 via the memory section 26.
[0116] As described above, the image displayed on the display
section 11 may become more easy to watch by rewriting the LUT 18 in
accordance with the brightness of the surroundings in which the
image display apparatus is located.
[0117] Moreover, the display image also becomes hard to watch in
case of an overall dark image or an overall bright image,
regardless of the surroundings of the image display apparatus. In
order to solve this, it may be arranged that the average value of
the signal level of the inputted video signal is calculated and the
LUT 18 may be rewritten so that the bit distribution of the display
image is rewritten to the average value.
[0118] In practice, the procedures are as follows: the average
value of the signal level of the video signal is calculated by the
average value calculating circuit 25. The control circuit 23 sets a
suitable value in accordance with the average value and rewrites
the LUT 18 via the memory section 26. In other words, the LUT 18
may be rewritten so that the bit distribution is rewritten in
accordance with the average value of the signal level of the video
signal that is an inputted signal.
[0119] Furthermore, in case of the color display image as described
previously, three units of the LUTs 18, which are data rewritable
in real time in accordance with various conditions as described
above, may be provided for the respective RGB colors, and are
adjusted to the respective colors for rewriting.
[0120] Note that, the example in the present embodiment is that an
8-bit digital signal is expanded to 10 bits and converted back to 8
bits again. However, the present invention is not limited to the
foregoing, and it may be applied to, for example, where a 10-bit
signal is displayed on a display of 10-bit display properties, or
where a 4-bit signal is displayed on a display of 4-bit display
properties, which is generally used in game machines or portable
phones.
[0121] As discussed earlier, the image processing apparatus of the
present invention is utilized for an image display apparatus
provided with display means of n-bit (n: a natural number) display
properties for displaying an image, upon receipt of a digital
signal as a video signal, in accordance with the digital signal.
The image processing apparatus is provided with a first signal
processing circuit and a second signal processing circuit. The
first signal processing circuit carries out gamma correction for an
n-bit digital signal inputted as the video signal, and converts the
digital signal into an m-bit (m>n, m: a natural number) digital
signal. The second signal processing circuit adds a noise signal
for reducing pseudo contour, to the m-bit digital signal from the
first signal processing circuit, then outputs a Q-bit (Q: a natural
number) digital signal, which is obtained by rounding off a less
significant (m-Q) bit (Q.ltoreq.n) from the m-bit digital signal,
to the above display means.
[0122] Accordingly, a display image of high quality with little or
no occurrence of the pseudo contour may be achieved by inputting an
n-bit digital signal into display means with display properties of
bit number less than or equal to n bit.
[0123] The image processing apparatus of the present invention is
provided with a signal processing circuit for outputting the Q-bit
digital signal, which is obtained by rounding off the less
significant (m-Q) bit (Q<m, m & Q are natural numbers) from
the m-bit digital signal after adding the noise signal for the
reduction of the pseudo contour into the inputted an m-bit digital
signal.
[0124] Therefore, it is not simply to omit the less significant
(m-Q) bit, but to achieve the expression of the display properties
equivalent to m bit in a pseudo manner by using a Q-bit digital
signal.
[0125] The noise is produced so that the average value of the
signal level of the noise signal to be added into the m-bit digital
signal is set to zero.
[0126] Accumulation of errors of signals caused by the noise signal
addition will not be resulted from adding a noise signal produced
as explained above into the digital signal in a random manner.
[0127] This may simplify the apparatus, thus offers a low-cost
image processing apparatus because means such as a memory and a
feedback circuit are no longer necessary, which are needed in the
conventional error diffusion method for preventing the accumulation
of errors when a noise signal is added into a digital signal. In
addition, it is also possible to include the entire circuit in the
source driving IC for greater economic effects.
[0128] The noise signal may have randomness with no regularity in
the cycle of the amplitude.
[0129] In this case, because no regularity of the cycle of the
amplitude of the noise signal interferes with the regularity of
reversing drives in case that an LCD apparatus is used as the image
display apparatus, the display quality of the display image may be
improved, accordingly.
[0130] The noise signal may, using an arbitrary noise pattern
table, be obtained by switching the starting point of the noise
pattern table per field or per noise pattern table.
[0131] In this case, because the noise level is evenly rounded down
by switching the starting point of the noise pattern table per
field, all pixels are allowed to express all the values of the
quantization levels.
[0132] Moreover, the quantization levels may be averaged per noise
pattern table by switching the starting point of the noise pattern
table per noise pattern table.
[0133] In an LCD apparatus, dot reversing, line reversing, or frame
reversing drives are generally carried out for AC drive, but the
cycle of the reversion is different for every LCD apparatus and is
not uniformed. Due to this reversing drive, a killer pattern may
exist in case there is regularity in the cycle of the amplitude of
the noise signal. In a rare case where the killer pattern is
displayed, a significant degradation in image quality is caused.
However, if there is no regularity in the cycle of the amplitude of
the noise signal, as described above, a killer pattern will never
exist and cause no degradation in image quality.
[0134] The histogram of the noise signal may be in Gaussian
dispersion where the zero value of the noise signal is set at the
center.
[0135] In this case, because the histogram of the amplitude of the
noise signal is in the Gaussian dispersion, the noise becomes least
visible, as the result, the display quality of the display image
can be improved.
[0136] The first signal processing circuit may be provided with bit
converting means for converting the inputted n-bit digital signal
into the m-bit digital signal in accordance with a value set in
advance.
[0137] In this case, the display quality of the display image can
be further improved by setting the value to be set in the bit
converting means, considering various conditions such as the
display properties of the display means or the environment
surrounding the image display apparatus.
[0138] For example, the value set in the bit converting means may
be rewritten so as to absorb the unevenness in the properties of
the drive means for driving the display means.
[0139] In this case, the unevenness among individual display means
need not be considered so strictly, because the unevenness in the
display properties generated for every display means can be
absorbed. As the result, the overall yield of the image display
apparatus is improved, thus significantly reducing the
manufacturing cost of the image display apparatus.
[0140] Besides, the value thus set by the bit converting means may
be rewritten in accordance with the brightness in the surroundings
of the image display apparatus.
[0141] In this case, the image can be displayed in high quality all
the time, regardless of the brightness of the surroundings of the
image display apparatus, by rewriting the value so that the tone
gradation of dark parts of the display image is made to be high
when the surroundings of the image display apparatus are dark,
while, when the surroundings of the image display apparatus are
bright, by rewriting the value so as to make that of bright part to
be high.
[0142] Furthermore, the value set by the bit converting means may
be rewritten in accordance with the display properties of an image
displayed on the display means.
[0143] In this case, the image can be displayed in high quality all
the time, regardless of the brightness of the image displayed on
the image display apparatus, by rewriting the value so as to make
the tone gradation of dark parts of the display image to be high
when the entire display image is dark, while, when the entire
display image is bright, by rewriting the value so as to make that
of bright part to be high.
[0144] The bit converting means may be realized by an LUT for
replacing the inputted signal with corresponding output value set
in advance.
[0145] In this case, the process in the bit converting means may be
promptly carried out because the output value varying depending on
with the inputted value is set in advance in the LUT. Additionally,
the entire arrangement of the image processing apparatus may be
streamlined by simplifying the structure of the bit converting
means.
[0146] Also, the bit converting means may be realized by a
calculating device for converting an n-bit digital signal into an
m-bit (m>n) digital signal by numerical calculation.
[0147] In this case, the calculating device can be a Central
Processing Unit (a CPU) or a Digital Signal Processor (a DSP).
[0148] Those CPU and DSP, which are programmable devices, have an
advantage to facilitate user interface for rewriting the value of
the bit converting means, thus making the operation easier.
[0149] When a color image is displayed on the image display
apparatus, the respective RGB colors may be provided with the first
signal processing circuit and the second signal processing
circuit.
[0150] Furthermore, the image display apparatus may be provided
with the image processing apparatus of the above arrangement.
[0151] In this case, the process of the signal inputted into the
display section of the image display apparatus is carried out in
the image processing apparatus. This eliminates the need of an
expensive apparatus for improving the display quality of the
display section, thus provides the image display apparatus of high
quality in a low cost.
[0152] The invention being thus described, it will be obvious that
the same way may be varied in many ways. Such variations are not to
be regarded as a departure from the spirit and scope of the
invention, and all such modifications as would be obvious to one
skilled in the art are intended to be included within the scope of
the following claims.
* * * * *